SEARCH · PubMed Health

Results for “Automated image segmentation”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 343 records · Page 19Linked to original sources

PubMed1998

Left-ventricle boundary detection from nuclear medicine images.

We present here a new algorithm for segmentation of nuclear medicine images to detect the left-ventricle (LV) boundary. In this article, other image segmentation techniques, such as edge detection and region growing, are also compared and evaluated. In the edge detection approach, we explored the relationship between the LV boundary characteristics in nuclear medicine images and their radial orientations: we observed that no single brightness function (eg, maximum of first or second derivative) is sufficient to identify the boundary in every direction. In the region growing approach, several criteria, including intensity change, gradient magnitude change, gradient direction change, and running mean differences, were tested. We found that none of these criteria alone was sufficient to successfully detect the LV boundary. Then we proposed a simple but successful region growing method--Contour-Modified Region Growing (CMRG). CMRG is an easy-to-use, robust, and rapid image segmentation procedure. Based on our experiments, this method seems to perform quite well in comparison to other automated methods that we have tested because of its ability to handle the problems of both low signal-to-noise ratios (SNR) as well as low image contrast without any assumptions about the shape of the left ventricle.

Heart Ventricles↗

PubMed2001

Computerized image analysis: estimation of breast density on mammograms.

An automated image analysis tool is being developed for the estimation of mammographic breast density. This tool may be useful for risk estimation or for monitoring breast density change in prevention or intervention programs. In this preliminary study, a data set of 4-view mammograms from 65 patients was used to evaluate our approach. Breast density analysis was performed on the digitized mammograms in three stages. First, the breast region was segmented from the surrounding background by an automated breast boundary-tracking algorithm. Second, an adaptive dynamic range compression technique was applied to the breast image to reduce the range of the gray level distribution in the low frequency background and to enhance the differences in the characteristic features of the gray level histogram for breasts of different densities. Third, rule-based classification was used to classify the breast images into four classes according to the characteristic features of their gray level histogram. For each image, a gray level threshold was automatically determined to segment the dense tissue from the breast region. The area of segmented dense tissue as a percentage of the breast area was then estimated. To evaluate the performance of the algorithm, the computer segmentation results were compared to manual segmentation with interactive thresholding by five radiologists. A "true" percent dense area for each mammogram was obtained by averaging the manually segmented areas of the radiologists. We found that the histograms of 6% (8 CC and 8 MLO views) of the breast regions were misclassified by the computer, resulting in poor segmentation of the dense region. For the images with correct classification, the correlation between the computer-estimated percent dense area and the "truth" was 0.94 and 0.91, respectively, for CC and MLO views, with a mean bias of less than 2%. The mean biases of the five radiologists' visual estimates for the same images ranged from 0.1% to 11%. The results demonstrate the feasibility of estimating mammographic breast density using computer vision techniques and its potential to improve the accuracy and reproducibility of breast density estimation in comparison with the subjective visual assessment by radiologists.

Biophysical Phenomena↗

PubMed1998

Ovarian ultrasound image analysis: follicle segmentation.

Ovarian ultrasound is an effective tool in infertility treatment. Repeated measurements of the size and shape of follicles over several days are the primary means of evaluation by physicians. Currently, follicle wall segmentation is achieved by manual tracing which is time consuming and susceptible to inter-operator variation. An automated method for follicle wall segmentation is reported that uses a four-step process based on watershed segmentation and knowledge-based graph search algorithm which utilizes priori information about follicle structure for inner and outer wall detection. The automated technique was tested on 36 ultrasonographic images of women's ovaries. Validation against manually traced borders has shown good correlation of manually defined and computer-determined area measurements (R2 = 0.85 - 0.96). The border positioning errors were small: 0.63+/-0.36 mm for inner border and 0.67+/-0.41 mm for outer border detection. The use of watershed segmentation and graph search methods facilitates fast, accurate inner and outer border detection with minimal user-interaction.

Algorithms↗

PubMed2004

Direct neural network application for automated cell recognition.

BACKGROUND: Automated cell recognition from histologic images is a very complex task. Traditionally, the image is segmented by some methods chosen to suit the image type, the objects are measured, and then a classifier is used to determine cell type from the object's measurements. Different classifiers have been used with reasonable success, including neural networks working with data from morphometric analysis. METHODS: Image data of cells were input directly into neural networks to determine the feasibility of direct classification by using pixel intensity information. Several types of neural network and their ability to work with cells in a complex patterned background were assessed for a variety of images and cell types and for the accuracy of classification. RESULTS: Inflammatory cells from animal biomaterial implants in rabbit paravertebral muscle were imaged in histologic sections. Simple, three-layer, fully connected, back-propagation neural networks and four-layer networks with two layers of a shared-weights neural network were most successful at classifying the cells from the images, with 97% and 98% correct recognition rates, respectively. CONCLUSIONS: The high accuracy recognition rate shows the potential for direct classification of visual image pixel data by neural networks.

Animals↗

PubMed1997

Verification of radiosurgery target point alignment with an electronic portal imaging device (EPID).

A procedure has been developed using an electronic portal imaging device (EPID) to verify that the center of a patient's lesion is aligned with the center of a treatment cone prior to treatment in a linac-based stereotactic radiosurgery procedure. The coordinates of the lesion center are set on the Brown-Roberts-Wells phantom base using a target simulator. A 3 mm tungsten ball, mounted on the target simulator, is used as the reference point for the planned isocenter. The target simulator is then attached to an adapter mounted on the linac couch, and an EPID image of the simulated target is acquired. The center of the circular-shaped radiation field is calculated from the centroid of the segmented EPID image, and the center of the tungsten ball is identified by an automated computer search algorithm. A summation filter is used to find the position of the lowest radiation intensity coincident with the center of the ball. The alignment error is defined as the difference between the center of the radiation field and the center of the ball. The accuracy of this method was tested and found to be within 0.2 mm. The advantage of the EPID-based procedure is that it can give quantitative offset values quickly for immediate readjustment. We have found that the method is also a convenient tool for testing room laser alignment and the accuracy of the treatment cones.

Algorithms↗

PubMed1999

Automated identification of stained cells in tissue sections using digital image analysis.

OBJECTIVE: To develop a novel automated image analysis system to differentiate immunohistochemically stained cells from background. STUDY DESIGN: Cell segmentation was performed by applying global thresholding algorithms to find an approximate threshold at which cells could be separated from background followed by a novel refinement algorithm to erode edge pixels of the region. To separate overlapping cells, a new decomposition method was developed that uses both semantic knowledge and high-level relational information. Both the cell segmentation and separation methods were evaluated on images of stained tissue sections and the manually outlined cell areas and numbers compared to the computed. RESULTS: Macrophage areas computed at the first stage by Otsu's algorithm did not differ significantly (P = .07) from those traced manually, while the areas computed by Kittler's and Kurita's algorithms did not agree (P < .01). Both Otsu's and Kurita's algorithms performed well when combined with edge pixel erosion. Kittler's algorithm proved unsuccessful even with edge erosion. Comparison of the computed and manually determined cell numbers showed a significant correlation, and regression analysis resulted in the unity curve. CONCLUSION: A combination of global thresholding and a novel edge erosion technique allowed identification of immunohistochemically stained macrophages; the computed cell areas agreed with the manual results.

Aorta↗

PubMed2004

Automated segmentation of necrotic femoral head from 3D MR data.

Segmentation of diseased organs is an important topic in computer assisted medical image analysis. In particular, automatic segmentation of necrotic femoral head is of importance for various corresponding clinical tasks including visualization, quantitative assessment, early diagnosis and adequate management of patients suffering from avascular necrosis of the femoral head (ANFH). Early diagnosis and treatment of ANFH is crucial since the disease occurs in relatively young individuals with an average age of 20-50, and since treatment options for more advanced disease are frequently unsuccessful. The present paper describes several new techniques and software for automatic segmentation of necrotic femoral head based on clinically obtained multi-slice T1-weighted MR data. In vivo MR data sets of 50 actual patients are used in the study. An automatic method built up to manage the segmentation task according to image intensity of bone tissues, shape of the femoral head, and other characters. The processing scheme consisted of the following five steps. (1) Rough segmentation of non-necrotic lesions of the femur by applying a 3D gray morphological operation and a 3D region growing technique. (2) Fitting a 3D ellipse to the femoral head by a new approach utilizing the constraint of the shape of the femur, and employing a principle component analysis and a simulated annealing technique. (3) Estimating the femoral neck location, and also femoral head axis by integrating anatomical information of the femur and boundary of estimated 3D ellipse. (4) Removal of non-bony tissues around the femoral neck and femoral head ligament by utilizing the estimated femoral neck axis. (5) Classification of necrotic lesions inside the estimated femoral head by a k-means technique. The above method was implemented in a Microsoft Windows software package. The feasibility of this method was tested on the data sets of 50 clinical cases (3000 MR images).

Automation↗

PubMed2005

The effects of early lead exposure on the brains of adult rhesus monkeys: a volumetric MRI study.

Little is known about direct effects of exposure to lead on central nervous system development. We conducted volumetric MRI studies in three groups of 17-year-old rhesus monkeys: (1) a group exposed to lead throughout gestation (n = 3), (2) a group exposed to lead through breast milk from birth to weaning (n = 4), and (3) a group not exposed to lead (n = 8). All fifteen monkeys were treated essentially identically since birth with the exception of lead exposure. The three-dimensional MRI images were segmented on a computer workstation using pre-tested manual and semi-automated algorithms to generate brain volumes for white matter, gray matter, cerebrospinal fluid, and component brain structures. The three groups differed significantly in the adjusted (for total brain size) volumes of the right cerebral white matter and the lateral ventricles. A significant reduction was noted in right cerebral white matter in prenatally exposed monkeys as compared to controls (p = 0.045). A similar reduction was detected in the white matter of the contralateral hemisphere; however, this difference did not achieve statistical significance (p = 0.143). Prenatally exposed monkeys also had larger right (p = 0.027) and left (p = 0.040) lateral ventricles. Depending on the timing of exposure during development, lead may exhibit differential effects with resultant life-long alterations in brain architecture.

Animals↗

PubMed2000

Frequency and clinical implications of fluid dynamically significant diffuse coronary artery disease manifest as graded, longitudinal, base-to-apex myocardial perfusion abnormalities by noninvasive positron emission tomography.

BACKGROUND: Diffuse coronary atherosclerosis is the substrate for plaque rupture and coronary events. Therefore, in patients with mild arteriographic coronary artery disease without significant segmental dipyridamole-induced myocardial perfusion defects, we tested the hypothesis that fluid dynamically significant diffuse coronary artery narrowing is frequently manifest as a graded, longitudinal, base-to-apex myocardial perfusion abnormality by noninvasive PET. METHODS AND RESULTS: In this study, 1001 patients with documented coronary artery disease by coronary arteriography showing any visible coronary artery narrowing underwent rest-dipyridamole PET perfusion imaging. Quantitative severity of dipyridamole-induced, circumscribed, segmental PET perfusion defects was objectively measured by automated software as the minimum quadrant average relative activity indicating localized flow limiting stenoses. Quantitative severity of the graded, longitudinal, base-to-apex myocardial perfusion gradient indicating fluid dynamic effects of diffuse coronary artery narrowing was objectively measured by automated software as the spatial slope of relative activity along the cardiac longitudinal axis. CONCLUSIONS: In patients with mild arteriographic disease without statistically significant dipyridamole-induced segmental myocardial perfusion defects caused by flow-limiting stenoses compared with normal control subjects, there was a graded, longitudinal, base-to-apex myocardial perfusion gradient significantly different from normal control subjects (P=0. 001) that was also observed for moderate to severe dipyridamole-induced segmental perfusion defects (P=0.0001), indicating diffuse disease underlying segmental perfusion defects; 43% of patients with or without segmental perfusion defects demonstrated graded, longitudinal, base-to-apex perfusion abnormalities beyond +/-2 SD of normal control subjects, indicating diffuse coronary arterial narrowing by noninvasive PET perfusion imaging.

Adult↗

PubMed1997

The use of active shape models for making thickness measurements of articular cartilage from MR images.

Previously reported studied to quantify articular cartilage have used labor-intensive manual or semi-automatic data-driven techniques, demonstrating high accuracy and precision. However, none has been able to automate the segmentation process. This paper describes a fast, automatic, model-based approach to segmentation and thickness measurement of the femoral cartilage in 3D T1-weighted images using active shape models (ASMs). Systematic experiments were performed to assess the accuracy and precision of the technique with in vivo images of both normal and abnormal knees. Segmentation accuracy was determined by comparing the results of the segmentation with the boundaries delineated by a radiologist. The mean error in locating the boundary was 0.57 pixels. To assess the precision of the measurement technique, the mean thickness of the femoral cartilage was calculated for repeated scans of five healthy volunteers. A mean coefficient of variation (CV) of 2.8% was obtained for the thickness measurements.

Cartilage, Articular↗

PubMed1989

A selective mapping algorithm for computer analysis of voided urine cell images.

One of the fundamental targets of the automated image analysis of cytologic preparations is the reduction of computer classification errors due to cells or other objects that do not lend themselves to image segmentation or that have morphologic features that may mislead the cell classification schemes. In prior work from this laboratory, the achievement of this goal was attempted by hierarchical analysis of sequential microscopic objects at high resolution. This paper reports on the successful development and implementation of an automated "selective mapping algorithm" that selects cells at low power for further analysis and eliminates a large proportion of unwanted "objects." The algorithm classifies the objects and extracts appropriate features from a 256 X 240 digital image obtained via a 10 X planachromatic objective. The five-node binary tree classifier used in this triage is described. The algorithm was trained and tested initially on 501 visually classified microscopic "objects," resulting in a correct acceptance rate of 61.3% and correct rejection rate of 81.3%. The selective mapping algorithm was subsequently integrated into the video-based image analysis system constructed at the Montefiore Medical Center for the diagnostic evaluation of sediments of voided urine. The algorithm was then tested on ten cytocentrifuge preparations for a preliminary evaluation of its performance. Up to 100 "objects" per case were selected by the algorithm for further classification by the computer at high power. Of the 810 "objects" selected by the selective mapping algorithm, 344 (42.5%) were classified by the computer at high resolution as cells of diagnostic value ("WELL" cells) and 466 were rejected.(ABSTRACT TRUNCATED AT 250 WORDS)

Algorithms↗

PubMed1997

Generalized region growing operator with optimal scanning: application to segmentation of breast cancer images.

Segmentation of medical images is a complex problem owing to the large variety of their characteristics. In the automated analysis of breast cancers, two image classes may be distinguished according to whether one considers the quantification of DNA (grey level images of isolated nuclei) or the detection of immunohistochemical staining (colour images of histological sections). The study of these image classes generally involves the use of largely different image processing techniques. We therefore propose a new algorithm derived from the watershed transformation enabling us to solve these two segmentation problems with the same general approach. We then present visual and quantitative results to validate our method.

Algorithms↗

PubMed2004

An improved segmentation method for in vivo microCT imaging.

UNLABELLED: Image segmentation methods for microCT can influence the accuracy of bone morphometry calculations. A new automated segmentation method is introduced, and its performance is compared with standard segmentation methods. The new method can improve the results of in vivo microCT, where the need to keep radiation dose low limits scan quality. INTRODUCTION: An important topic for microCT analysis of bone samples is the segmentation of the original reconstructed grayscale data sets to separate bone from non-bone. Problems like noise, resolution limitations, and beam-hardening make this a nontrivial issue. Inappropriate segmentation methods will reduce the potential power of microCT and may introduce bias in the architectural measurements, in particular, when new in vivo microCT with its inherent limitations in scan quality is used. Here we introduce a new segmentation method using local thresholds and compare its performance to standard global segmentation methods. MATERIAL AND METHODS: The local threshold method was validated by comparing the result of the segmentation with histology. Furthermore, the effect of choosing this new method versus standard segmentation methods using global threshold values was investigated by studying the sensitivity of these methods to signal to noise ratio and resolution. RESULTS: Using the new method on high-quality scans yielded accurate results and virtually no differences between histology and the segmented data sets could be observed. When prior knowledge about the volume fraction of the bone was available the global threshold also resulted in appropriate results. Degrading the scan quality had only minor effects on the performance of the new segmentation method. Although global segmentation methods were not sensitive to noise, it was not possible to segment both lower mineralized thin trabeculae and the higher mineralized cortex correctly with the same threshold value. CONCLUSION: At high resolutions, both the new local and conventional global segmentation methods gave near exact representations of the bone structure. When scanned samples are not homogenous (e.g., thick cortices and thin trabeculae) and when resolution is relatively low, the local segmentation method outperforms global methods. It maximizes the potential of in vivo microCT by giving good structural representation without the need to use longer scanning times that would increase absorption of harmful X-ray radiation by the living tissue.

Algorithms↗

PubMed2003

Three-dimensional ultrasound imaging and its use in quantifying organ and pathology volumes.

Although ultrasonography is an important cost-effective imaging modality, technical improvements are needed before its full potential is realized for accurate and reproducible monitoring of disease progression or regression. Two-dimensional viewing of three-dimensional anatomy, using conventional ultrasonography, limits our ability to quantify and visualize pathology and is partly responsible for the reported variability in diagnosis and monitoring of disease progression. Efforts of investigators have focused on overcoming these deficiencies by developing 3D ultrasound imaging techniques that are capable of acquiring B-mode images using existing conventional ultrasound systems, reconstructing the information into 3D images, and then allowing interactive viewing of the 3D images on inexpensive desktop computers. In addition, the availability of 3D ultrasound images has allowed the development of manual and semi-automated techniques to quantify normal and abnormal anatomical volumes. In this paper we review our semi-automated 3D segmentation approaches for segmenting the surface of the carotid arteries and plaques, and segmenting the prostate. These techniques demonstrate that efficient segmentation techniques can be used with 3D ultrasound images to quantify anatomical organ volumes and morphology.

Carotid Arteries↗

PubMed2000

Automatic morphology-based brain segmentation (MBRASE) from MRI-T1 data.

A method called morphology-based brain segmentation (MBRASE) has been developed for fully automatic segmentation of the brain from T1-weighted MR image data. The starting point is a supervised segmentation technique, which has proven highly effective and accurate for quantitation and visualization purposes. The proposed method automates the required user interaction, i.e., defining a seed point and a threshold range, and is based on the simple operations thresholding, erosion, and geodesic dilation. The thresholds are detected in a region growing process and are defined by connections of the brain to other tissues. The method is first evaluated on three computer simulated datasets by comparing the automated segmentations with the original distributions. The second evaluation is done on a total of 30 patient datasets, by comparing the automated segmentations with supervised segmentations carried out by a neuroanatomy expert. The comparison between two binary segmentations is performed both quantitatively and qualitatively. The automated segmentations are found to be accurate and robust. Consequently, the proposed method can be used as a default segmentation for quantitation and visualization of the human brain from T1-weighted MR images in routine clinical procedures.

Adolescent↗

PubMedDate not supplied

3D visualization, analysis, and treatment of the prostate using trans-urethral ultrasound.

In the year 2000, it is estimated that over 20,000 men underwent transperineal interstitial permanent prostate brachytherapy (TIPPB) for treatment of prostate cancer. Trans-urethral ultrasound (TUUS) is a new interactive, real-time 3D imaging method that may be effective in therapy-guidance during and after TIPPB. TUUS provides higher resolution than trans-rectal ultrasound (TRUS). TUUS can be used to accurately localize radioactive seeds and therefore contribute to more accurate determination of radiation dose distribution throughout the tissue after the completion of the procedure, similar to information currently provided by expensive and offline CT scans. A TUUS catheter can be used to acquire 2D section images or 3D volume images for detailed analyses of the prostate and associated tissue. Initial development of TUUS imaging was carried out on an ultrasound-equivalent prostate phantom with cylindrical dummy radiation sources. This was followed by preliminary studies in animals and then in patients. Both CT and TRUS data were acquired in these studies for comparative purposes. Segmentation of the prostate capsule and radioactive seeds was carried out using several semi-automated 3D algorithms and image processing techniques. Presentation of the data to the clinician is provided by a variety of complementary 2D and 3D display methods. In comparison with the CT data, TUUS data provided both greater spatial resolution and better soft tissue differentiation. In comparison to the TRUS data, TUUS data provided greater resolution and better seed localization. Combining these advantages suggests the possibility of TUUS becoming the exclusive imaging method in prostate cancer brachytherapy.

Humans↗

PubMedDate not supplied

An automated image cytometry system for monitoring DNA ploidy and other cell features of radiotherapy and chemotherapy patients.

DNA content and distribution in cell nuclei were studied in samples of fine-needle aspiration (FNA) from 27 locally advanced breast and head and neck cancers in two going randomized trials that compared accelerated fractionation to standard fractionation radiation in locally advanced breast cancer and head and neck cancer. Two image cytometry methods were compared: a new, fully automated DNA image cytometry system (AIC) and a conventional image cytometry (CIC) system with manual selection, focusing, and segmentation of cells. The results of both techniques were compared on the basis of DNA histogram parameters including DNA index (DI), mean DNA values (MDV), and Auer's DNA histogram patterns. An excellent correlation was achieved between the two imaging techniques in terms of DI (r=0.985, p<0.001) and MDV (r=0.951, p<0.001) as well as between Auer's histogram patterns, where both methods agreed completely. It was concluded in these analyses that the two image cytometry methods were equivalent. However, the AIC offered an advantage by scanning samples in a fully automated way, which represented significant time saving for cytopathologists working with the system, as well as a larger number of cells used in the automated analysis. With the automated image cytometer, 500 relevant cells were collected and analyzed in about 10 minutes, where with the interactive (manual) method, it took typically an hour to collect and analyze only about 250 cells. Seventeen samples were sufficient for flow analysis. Image cytometry and flow cytometry showed good agreement in DI determination; however, three cases reported as diploid by flow cytometry were found to be aneuploid by image cytometry techniques.

Antineoplastic Combined Chemotherapy Protocols↗

PubMed1989

Evaluation of automated threshold selection methods for accurately sizing microscopic fluorescent cells by image analysis.

The accurate measurement of bacterial and protistan cell biomass is necessary for understanding their population and trophic dynamics in nature. Direct measurement of fluorescently stained cells is often the method of choice. The tedium of making such measurements visually on the large numbers of cells required has prompted the use of automatic image analysis for this purpose. Accurate measurements by image analysis require an accurate, reliable method of segmenting the image, that is, distinguishing the brightly fluorescing cells from a dark background. This is commonly done by visually choosing a threshold intensity value which most closely coincides with the outline of the cells as perceived by the operator. Ideally, an automated method based on the cell image characteristics should be used. Since the optical nature of edges in images of light-emitting, microscopic fluorescent objects is different from that of images generated by transmitted or reflected light, it seemed that automatic segmentation of such images may require special considerations. We tested nine automated threshold selection methods using standard fluorescent microspheres ranging in size and fluorescence intensity and fluorochrome-stained samples of cells from cultures of cyanobacteria, flagellates, and ciliates. The methods included several variations based on the maximum intensity gradient of the sphere profile (first derivative), the minimum in the second derivative of the sphere profile, the minimum of the image histogram, and the midpoint intensity. Our results indicated that thresholds determined visually and by first-derivative methods tended to overestimate the threshold, causing an underestimation of microsphere size. The method based on the minimum of the second derivative of the profile yielded the most accurate area estimates for spheres of different sizes and brightnesses and for four of the five cell types tested. A simple model of the optical properties of fluorescing objects and the video acquisition system is described which explains how the second derivative best approximates the position of the edge.

Cyanobacteria↗